Multi Scale Identity-Preserving Image-to-Image Translation Network for Low-Resolution Face Recognition

TL;DR

This paper introduces a deep neural network that super-resolves low-resolution faces into high-resolution images while preserving identity, significantly improving face recognition accuracy in low-resolution scenarios like surveillance.

Contribution

The proposed end-to-end identity-preserving image-to-image translation network is novel in jointly super-resolving and maintaining identity information across multiple scales.

Findings

Outperforms existing super-resolution methods in identity preservation.

Achieves higher face recognition accuracy on low-resolution datasets.

Effective on unseen identities and natural low-resolution images.

Abstract

State-of-the-art deep neural network models have reached near perfect face recognition accuracy rates on controlled high-resolution face images. However, their performance is drastically degraded when they are tested with very low-resolution face images. This is particularly critical in surveillance systems, where a low-resolution probe image is to be matched with high-resolution gallery images. super-resolution techniques aim at producing high-resolution face images from low-resolution counterparts. While they are capable of reconstructing images that are visually appealing, the identity-related information is not preserved. Here, we propose an identity-preserving end-to-end image-to-image translation deep neural network which is capable of super-resolving very low-resolution faces to their high-resolution counterparts while preserving identity-related information. We achieved this by training a very deep convolutional encoder-decoder network with a symmetric contracting path between corresponding layers. This network was trained with a combination of a reconstruction and an identity-preserving loss, on multi-scale low-resolution conditions. Extensive quantitative evaluations of our proposed model demonstrated that it outperforms competing super-resolution and low-resolution face recognition methods on natural and artificial low-resolution face data sets and even unseen identities.